Rice Monitoring and Yield Estimation in Dumangas, Iloilo, Philippines Using Satellite Imagery

Abstract

An agricultural monitoring procedure utilizing the combination of remotely sensed data, vegetation indices, ground truth information, and local knowledge on the phenology and development of rice, which is the staple crop produced in Dumangas, Iloilo, Philippines, was established and analyzed in this study. The areas with standing crops and their corresponding growth stage were reported for the 2nd and 3rd cropping of 2020. The spatio-temporal analysis determined that NDVI is directly correlated with the photosynthetic activity and development of the crop. Dry season cropping lasted from October to January, while the 3rd cropping lasted from February to April. The percentage of the rice areas planted per month were also studied and observed that the area spatially and numerically increased as farmers planted during the start of the season. The ground data from the 20 GCPs were compared with the Sentinel 2 NDVI. The phenological metrics for each point characterized the development of the crop. From the investigation, the municipality commonly used hybrid varieties, which have a growth duration of 85 to 100 days. The NDVI ranges were 0.163 to 0.309, 0.35 to 0.682, 0.535 to 0.671, and 0.235 to 0.533 for seedling, vegetative, reproductive, and ripening. APSIM was used for yield modeling of the cropping season 2019 and observed to have a good correlation during the wet season. Underestimation was detected for cropping seasons during times with limited rainfall. This study was beneficial for the LGU in crop estimation as it produced highly reliable spatial and temporal results.